Treating periprosthetic joint infection: adopting an individualized approach

How should a clinician tackle the management and treatment of a periprosthetic joint infection (PJI)? An individualized approach must be developed for each patient, one that integrates appropriate surgical intervention and antimicrobial strategies. The complexities of treating a biofilm infection calls for careful consideration of each case, addressing the acute or chronic nature of infection.

Compared to primary arthroplasty and revision for aseptic loosening, surgical revision for PJI has significantly higher associated costs and higher demands on physician resources [1]. A study of 465,209 hip revisions in the US between 2003 and 2013 found the PJI prevalence to be 15% with hospitalization costs of USD 31,529 [2]. When indirect treatment costs, such as lost wages, were taken into account, another estimate determined the base cost of treating an infected total hip arthroplasty (THA) to be between USD 389,307 and USD 474, 004, depending on the age of the patient at the time of infection [3].

See Part 1 of this article series for further elaboration of PJI risk factors and pre-, intra-, and postoperative infection prevention strategies. Part 2 looks at different components of diagnosing PJI, culminating in a recommended diagnostic algorithm.

Treatment goals

The goal in treating a PJI is to deliver a pain-free and functional joint, which can be achieved by eliminating the infection [4]. Treatment plans should be customized to each individual patient and contingent on the microorganism(s) responsible for the infection [5]. These are simple statements to make yet the pernicious nature of device biofilm, coupled with antibiotic resistance, can make a conclusive diagnosis and successful treatment a challenge.

 
Treatment success rates

Compared to hospital admissions for aseptic loosening, PJI patients have a 2-times higher chance of in-hospital mortality. Treatment of PJI cases often requires multiple surgical admissions and the risk of mortality accumulates with each surgery [6]. Infection of implanted prosthetic joints is more often associated with revisions than with primary arthroplasty [7].

The rate of PJI eradication varies. Sidhu et al reported an overall rate of 50% after two years and 38.9% after five, pointing out that if there is a polymicrobial infection that includes a fungal component, patients are less likely to have their infection cured [8]. A fungal PJI has lower treatment success rates [9]. Akgun et al were able to obtain a 3 year infection-free survival rate of 89.3% in 84 hip PJI patients who underwent two-stage prosthesis exchange [10]. Similarly, Aboltins et al reported a 2 year infection-free survival of 87% in 41 hip and knee PJI patients [11].

Both abovementioned studies conducted patient follow up for 3 and 2 years, respectively. It has been suggested that follow up of any less than 10 years will not capture late PJI, leading to underreporting of its occurrence [7].

Treatment options: two elements—surgery and antibiotics

Depending on how much time has passed from implantation, or the length of symptoms, PJI is classified as either acute or chronic. This designation also indicates the maturity stage of the pathogenic biofilm, the virulence of the microorganism(s), and is associated with different clinical features [12]. See Table 1. This classification also provides direction on how to begin a treatment plan, which has two interrelated components—surgery and antibiotics. Let’s look at surgical options first.

 
Table 1. Classification of prosthetic joint infection (PJI) and associated surgical treatment option(s)

^a Fosfomycin-sodium is preferred over fosfomycin-calcium due to better mechanical properties of PMMA.

^b Available as colistin-sodium or colistin-sulfate (equal efficacy).

^c Improved efficacy and antimicrobial release in combination with gentamicin 1 g and clindamycin 1 g, which can be used as basis for admixing additional antimicrobials.

^d These AM concentrations do not fulfill the mechanical ISO requirements for fixation cement.

^e Literature is still controversial regarding minimal effective concentrations.

General considerations:

• When additional antimicrobials are admixed, industrially impregnated cements are preferred over plain cements (better mechanical properties and elution due to synergistic release).
• Antimicrobial susceptibility testing results are applicable for systemic antimicrobial application and might not be valid for local antimicrobial application due to high local concentrations and synergistic activity.
• Side effects and interactions of local antimicrobials are rare. However, serum concentrations of vancomycin and gentamicin should be monitored in patients with kidney insufficiency (eGFR <60ml/min) and/or intravenous application.
• Only use sterile antimicrobials in powder form. Liquid antimicrobials are not recommended due to inhomogeneous distribution in PMMA. Antibiotics that interfere with polymerization process (rifampin or metronidazol) or which are thermolabile or sensitive to oxidation (eg, some beta lactams) should not be used.
• Data on mechanical stability are not available for combinations of >2 antimicrobials. If possible, the total amount of antimicrobials should not exceed 10% of the PMMA powder weight (= 4 g per 40 g).
• Recommendations are based on studies with PALACOS®/COPAL® PMMA cements and literature data. Elution data depend on the PMMA cement basis used.
• Do not use vacuum mixing for preparation of spacer cement (higher porosity -> better antimicrobial elution).
• Video-tutorial about admixing of antimicrobials in bone cement is available.

The type of microorganism(s) causing the PJI will dictate if a one- or two-stage exchange should be attempted—difficult to treat (DTT) infections need a prolonged interval without the prosthesis present for enhanced antimicrobial treatment and a two-stage is better suited to this situation [12].

A 2 (short) to 6 week (long) interval before reimplantation is recommended [38]. The timing will be patient specific and reflect the type of pathogen(s) identified. More than eight weeks between procedures is not advised because antimicrobial effects of spacers and cement wane, and by this time are below inhibitory concentrations [12]. See Figure 1 for a schematic of surgical options and treatment timing.

 
Three-stage exchange

During the interval between prosthesis removal and reimplantation, certain patients may display signs of persistent or DTT infection, lingering wound discharge, or have a fungal element to their PJI. In such cases, a three-stage exchange may be required. This involves the insert, then after a period of time (generally 3 weeks), the removal and replacement of antibiotic-loaded spacers as well as additional debridement to further decrease the bacterial load, then another 3 weeks before the prosthesis is replaced. See Figure 1 for a schematic of surgical options and treatment timing.

 
If it doesn’t work

Occasionally, second or even tertiary two-stage revisions are performed if the infection persists after the first. However, each revision procedure is associated with higher failure and mortality rates, particularly in patients with DTT or resistant bacteria, an inadequate soft-tissue envelope, and poor general health [6, 42]. Rarely, PJI cannot be resolved. Options in these exceptional cases are: amputation, permanent removal of the prosthesis, creation of an iatrogenic stable sinus tract, and in the case of the knee, arthrodesis [12, 42, 43].
 

 
Antimicrobial treatment recommendations

Considering that biofilm bacteria are 100–1000 times more resistant to antibiotics than planktonic bacteria, considerably higher antibiotic concentrations (10–8000 fold higher) are needed to eliminate biofilms [44, 45].

Protecting a new implant from colonization and the development of biofilm is important. If mobile parts are being exchanged, a one-stage exchange is planned, or the prosthesis is being reimplanted, antibiotic prophylaxis is recommended 30–60 minutes before skin incision or tourniquet closure [12].

Whereas there is general agreement concerning the prevention and diagnosis of PJI, considerable differences in option exist around the antimicrobial treatment of PJI. “Controversies concerning antibiotics include the choice and duration of systemic antibiotic treatment and indications for admixing local antibiotics in bone cement, both for permanent fixation and temporary spacer,” says Trampuz.

Current recommendations for local antimicrobial therapy are summarized in the side box on PMMA. See Table 3 below for current recommendation for systemic antibiotics. According to Trampuz, “It remains controversial whether application of antimicrobial powder in surgical cement is beneficial or harmful. In contrast, there is increasing evidence supporting the use of antimicrobial-coated implants in high-risk patients.”

Developing antibiotic resistance in pathogens is a serious concern and it is recommended to not start broad-spectrum antimicrobial treatment until debridement and initial intravenous (IV) antibiotics are completed [12]. See Figure 1 for further clarification of the relationship between antibiotic delivery and surgical procedure timing. Table 3 is a more comprehensive resource that matches the most appropriate antibiotic with a target microorganism, suggesting an appropriate dose and dosing pathway.

Identification of the microorganisms is critical to matching treatment with the most effective antibiotic. This is done in cooperation with the microbiologist and is informed by susceptibility testing. Once the infecting agent(s) are determined, then targeted IV therapy should begin. Switching to oral antibiotics is typically performed 1 to 2 weeks after surgery if [12]:

• An oral antibiotic with good bone penetration is available
• Wounds are dry
• Local conditions are satisfactory
• Serum C-reactive protein (CRP) levels (almost) normalizes/declines significantly

Izakovicova et al emphasize that bio-film-active antibiotics should not be used when a spacer is in place (during a two- or three-stage exchange) and only started when “the definitive prothesis is implanted, wounds are dry, and drains are removed” [12].

 
Suppressive antibiotic treatment

For various reasons some patients, particularly the elderly, are not able to support the removal of their infected prosthesis, even if it is indicated. Although it encourages the development of antibiotic resistance (in 23% of patients in one study) [46], suppressive antibiotic treatment is considered a long-term option [47–49]. If the microorganism is drug resistant, long-term suppressive antibiotic treatment of over a year, or even life long, may be indicated in certain cases [38]. To prevent development of resistance, suppression should preferably be applied only after surgery was performed to reduce bacterial load. See Table 3 for further information.

 
Antibiotic holidays: yes or no? No

“Antibiotic holidays”, stopping antibiotic administration before re-implantation, is no longer recommended [12]. Patients are more likely to experience a flare of infection when on a drug holiday [50], and may relapse after reimplantation. Using perioperative antibiotic prophylaxis before incision does not appear to negatively influence the diagnostic sensitivity of intraoperative biopsies taken during exchange of mobile parts, one-stage exchange or reimplantation surgery [51, 52].

 
Table 3. Recommended antimicrobial treatment

^a Total duration of therapy: 12 weeks, usually 2 weeks intravenously, followed by oral route

^b Laboratory testing 2x weekly: leukocytes, CRP, creatinine / eGFR, liver enzymes (AST/SGOT and ALT/SGPT). Dose-adjustment according to renal function and body weight (< 40 kg or > 100 kg)

^c Penicillin allergy of NON-type 1 (eg, skin rash): cefazolin (3 x 2 g IV). In case of anaphylaxis(= type 1-allergy such as Quincke´s edema, bronchospasm, anaphylactic shock) or cephalosporin allergy: vancomycin (2 x 1 g IV) or daptomycin (1 x 8 mg/kg IV) Ampicillin/sulbactam is equivalent to amoxicillin/clavulanic acid (3 x 1.2 g IV)

^d Rifampin is administered only after the new prosthesis is implanted. Add it already to intravenous treatment as soon as wounds are dry and drains removed; in patients aged > 75 years, rifampin is reduced to 2 x 300 mg PO

^e Check Vancomycin through concentration (take blood before next dose) at least 1x/week; therapeutic range: 15-20 µg/ml

^f Give only, if gentamicin high-level (HL) is tested susceptible (consult the microbiologist). In gentamicin HL-resistant E. faecalis: gentamicin is exchanged with ceftriaxone (1 x 2 g IV)

^g Add IV treatment (piperacillin/tazobactam 3 x 4.5 g or ceftriaxone 1 x 2 g or meropenem 3 x 1 g IV) in the first postoperative days (until wound is dry)

h After loading dose (70 mg on day 1, reduce to 50 mg in patients weighing < 80 kg from day 2)

 
Abbreviation: IV, intravenous; PO, per os.

 
Used with permission. From: Pocket Guide to Diagnosis & Treatment of PJI; PRO-IMPLANT Foundation (October 2019).

PRO-IMPLANT Foundation recommended treatment algorithm

Figure 2 condenses everything we have discussed into a simple schematic that leads you through the decision-making process for treating acute and chronic PJI. The PRO-IMPLANT Foundation is a non-profit education organization with the sole aim of the better understanding diagnosis and treatment of biofilm and implant-related infections and therewith improve treatment outcomes and patients life quality worldwide.

New innovations—the future looks bright

There is hope that more effective measures to combat PJI are coming. Borens is optimistic about developments on the horizon that could improve prevention, diagnosis, and treatment options. “We are living in an exciting time and we are adopting more improvements every year. But this advancement is necessary as the number of infected joints is going to increase enormously in the years to come. There is lots of research being done on improving diagnostics with new tests. There are projects on coatings for implants and on local treatment of infection. Finally, there is research being done on new antimicrobials like phages.”

Trampuz also highlights phages as a promising development. “Bacteriophage therapy is a rapidly evolving alternative to antibiotics, in particular when multiresistant bacteria and biofilms are involved. Bacteriophages are viruses, which specifically infect and destroy bacteria within hours. When applied locally during or after surgery, they have the potential to cure infections. Future research focuses in appropriate phage selection, concentration, administration and risk of resistance development.”

Innovative strategies are being explored and the most promising advancements could very well expand the surgeon’s options for preventing, diagnosing, and treating PJI—here are just a few, see Table 4 [53, 54].

 
Table 4. A selection of promising innovative technologies that could impact the occurrence, prevention, diagnosis and/or treatment of PJI in the future

 
We must also mention the advances that are being made to assist in the diagnosis of PJI. Trampuz is enthusiastic about “novel biomarkers that are directly detecting pathogens or their metabolites. These may have better sensitivity and specificity than current biomarkers detecting inflammatory host response (such as synovial fluid leukocytes, alpha-defensin, calprotectin, interleukin-6). An example of this type of marker is D-lactate, which is exclusively produced by bacteria and can be easily measured in body fluids.”

Conclusion

The diagnosis and treatment of PJI can be complex and require the collaboration of a multidisciplinary health care team. Knowing and managing the risks factors that predispose arthroplasty patients to PJI can go a long way in minimizing the chances for infection [72], as well as implementing improved infection control procedures and hygiene protocol adherence. But if a PJI does take hold, evidence-based algorithms and international recommendations for treatment offer great guidance toward resolving the problem.

Above all, Borens reminds us that even though there are accepted protocols in use around the world, each patient is different and will require a customized approach to diagnosis and treatment. “Most of us surgeons have similar ideas for surgical treatment, and more and more of us are looking to well established protocols for antibiotic treatment. But of course, we must keep in mind that there will always be certain special situations when you will be forced to think out of the protocol to adapt it to a special patient’s situation.”